Risk of Injurious Fall and Hip Fracture up to 26 y before the Diagnosis of Parkinson Disease: Nested Case-Control Studies in a Nationwide Cohort

Abstract

Background: Low muscle strength has been found in late adolescence in individuals diagnosed with Parkinson disease (PD) 30 y later. This study investigated whether this lower muscle strength also may translate into increased risks of falling and fracture before the diagnosis of PD.

Methods and findings: Among all Swedish citizens aged ≥50 y in 2005, two nested case-control cohorts were compiled. In cohort I, individuals diagnosed with PD during 1988-2012 (n = 24,412) were matched with up to ten controls (n = 243,363), and the risk of fall-related injuries before diagnosis of PD was evaluated. In cohort II, individuals with an injurious fall in need of emergency care during 1988-2012 (n = 622,333) were matched with one control (n = 622,333), and the risk of PD after the injurious fall was evaluated. In cohort I, 18.0% of cases and 11.5% of controls had at least one injurious fall (p < 0.001) prior to PD diagnosis in the case. Assessed by conditional logistic regression analysis adjusted for comorbid diagnoses and education level, PD was associated with increased risks of injurious fall up to 10 y before diagnosis (odds ratio [OR] 1.19, 95% CI 1.08-1.31; 7 to <10 y before diagnosis) and hip fracture ≥15 y before diagnosis (OR 1.36, 95% CI 1.10-1.69; 15-26 y before diagnosis). In cohort II, 0.7% of individuals with an injurious fall and 0.5% of controls were diagnosed with PD during follow-up (p < 0.001). The risk of PD was increased for up to 10 y after an injurious fall (OR 1.18, 95% CI 1.02-1.37; 7 to <10 y after diagnosis). An important limitation is that the diagnoses were obtained from registers and could not be clinically confirmed for the study.

Conclusions: The increased risks of falling and hip fracture prior to the diagnosis of PD may suggest the presence of clinically relevant neurodegenerative impairment many years before the diagnosis of this disease.

Fig 1. Selection process of the nested case—control cohorts.

Fig 2. Overview of data collection and study times.

(A) The timeline of data collection according to calendar dates. (B) Retrospective analyses in cohort I. The gray arrows provide a schematic illustration of single individuals’ study time, with events (fall/hip fracture) marked (*) above the line. Left censoring (†) occurs at the beginning of the data collection time. Time intervals used in analyses are indicated by horizontal lines (0 to <3 mo, 3 to <12 mo, 1 to <2 y, 2 to <3 y, 3 to <4 y, 4 to <5 y, 5 to <7 y, 7 to <10 y, 10 to <15 y, and 15 to 26 y before the index date). (C) Prospective analyses in cohort II. Individual study times are indicated the same way as in (B), with the follow-up time split into the following intervals after the index date: 0 to <3 mo, 3 to <12 mo, 1 to <2 y, 2 to <3 y, 3 to <4 y, 4 to <5 y, 5 to <7 y, 7 to <10 y, 10 to <15 y, and 15 to 25 y. Marks above the arrow indicate PD (*) or censoring (†) due to death or end of data collection.

Fig 3. Hazard ratio for Parkinson disease according to fall in cohort II.

Hazard ratio for PD after first injurious fall in cohort II, estimated by a flexible parametric Royston—Parmar model adjusted for sex, year of birth, education level, and comorbid diagnoses (dementia, stroke, myocardial infarction, diabetes mellitus, depression, alcohol dependency or abuse, drug dependency or abuse). The gray areas represent the 95% confidence intervals.